LLM Agent Research Scientist 大模型智能体研究科学家

Define and lead frontier research for LLM-powered agents, spanning agent architectures, reasoning and planning, tool use, and multi-agent frameworks. Drive post-training and RL innovation to build reliable agent capabilities, including reward modeling, synthetic data, and environment design. Create rigorous benchmarks for correctness, safety, robustness, and real-world task success, and partner closely with engineering, product, design, and data teams to ship research into core Canva experiences.